01-Data-Preprocessing.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CS F415 - Data Mining \n",
    "\n",
    "## Section 1 - Data Preprocessing \n",
    "\n",
    "This notebook will contain basic information about the various data preprocessing techniques as mentioned in the lectures followed by code snippets in python that will help you realise these methods on a sample dataset. The Notebook will first contain brief information about the types of data and then cover the various preprocessing techniques. The preprocessing techniques that will be covered in this notebook are:\n",
    "* Aggregation\n",
    "* Sampling\n",
    "* Dimensionality Reduction \n",
    "* Feature Subset Selection\n",
    "* Feature Creation Discretization and Binarization\n",
    "* Attribute Transformation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 1: Python and Jupyter Environment Setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This section is for people who have never worked with Jupyter or used Python for Data Science Related Applications. In case you already have the environment setup, feel free to skip to this [section](#cell2)\n",
    "\n",
    "Check out: [What are Jupyter Notebooks/IPython Notebooks?](https://www.youtube.com/watch?v=jZ952vChhuI)\n",
    "\n",
    "In order to run these notebooks on your device or to create notebooks such as these for hands on experience, there are several approaches:\n",
    "\n",
    "#### Approach 1. Google Colab \n",
    "\n",
    "Advantages: \n",
    "* Google Colab allows you to remotely access a computer on the cloud which already has the environment setup and the most popular Machine Learning (ML), Deep Learning (DL) and Natural Language Processing (NLP) Libraries installed which allows a user with any hardware to perform computationally-intensive ML and DL tasks even with weak hardware\n",
    "* This means you can directly start getting some hands on experience with these notebooks.\n",
    "* For Beginners with little experience in setting up environments, Google Colab is the best way to start. \n",
    "* You have access to more powerful Server-Class Graphics Processing Units (GPUs) as compared to the ones available to general consumers which as Desktop-Class GPUs which are sometimes required for more powerful Machine Learning and Deep Learning Tasks\n",
    "\n",
    "Disadvantages: \n",
    "* The Cloud environment means that the resources that google can provide for free are limited and thus, upon some idle time away from the notebook your session would be reset including the storage and any files you may have saved in that session. \n",
    "* The GPUs that are available are often available for limited periods of time and are easy to exhaust. If exhausted, one may need to purchase additional GPU credits.\n",
    "\n",
    "[Link to Google Colab]('https://colab.research.google.com/')\n",
    "\n",
    "---\n",
    "\n",
    "#### Approach 2. The next method is to use Anaconda Navigator and a Local Jupyter Server. \n",
    "\n",
    "This method requires a bit of initial setup and this varies based on the Operating System of the user and this method is hardware dependent.\n",
    "\n",
    "Advantages:\n",
    "* The server is always running as long as you decide to kill it which means that all the local variables and the files you require for input and any output files generated will always be safely stored on your local storage.\n",
    "* Running locally means that your data remains private and lets you access the notebook when offline\n",
    "\n",
    "Disadvantages:\n",
    "* The initial setup with anaconda could be error prone and tricky if the user already has python installed on their system. \n",
    "* Forgetting to kill the server before quitting the jupyter environment may result in some issues. \n",
    "* The hardware a user can use is limited to the hardware that a user can access\n",
    "\n",
    "[Setup Tutorial for Windows](https://www.xda-developers.com/how-install-jupyter-notebook-windows/)\n",
    "\n",
    "[Setup Tutorial for MacOS](https://www.youtube.com/watch?v=A_xY5eL_IKU)\n",
    "\n",
    "[Setup Tutorial for Linux(Debian Based Distros - Ubuntu)](https://www.youtube.com/watch?v=wotkLCoJ_ks)\n",
    "\n",
    "---\n",
    "\n",
    "#### Approach 3. Installing Python and Setting up Jupyter Environment on Visual Studio Code\n",
    "\n",
    "Advantages:\n",
    "* Similar to Jupyter Server, it runs locally so it contains the same advantages\n",
    "* Since it runs on a fully fledged IDE like Visual Studio Code, One can customize the interface to fit their convenience and install additional extensions which could help the experience. \n",
    "* Server is always killed when you close Visual Studio Code\n",
    "* Similar Setup for all operating systems once you get Python Setup and install Visual Studio Code\n",
    "\n",
    "Disadvantages:\n",
    "* Restriced to user hardware\n",
    "* Initial Setup can be a bit tedious\n",
    "\n",
    "Setup Tutorial for Python: \n",
    "* [Windows]('https://docs.python.org/3/using/windows.html')\n",
    "* [MacOS]('https://docs.python.org/3/using/mac.html')\n",
    "* [Linux (Ubuntu)]('https://phoenixnap.com/kb/how-to-install-python-3-ubuntu')\n",
    "\n",
    "[Setup Tutorial for Visual Studio Code]('https://code.visualstudio.com/learn/get-started/basics')\n",
    "\n",
    "[Setup Tutorial for Jupyter on Visual Studio Code]('https://code.visualstudio.com/docs/datascience/jupyter-notebooks')\n",
    "\n",
    "---\n",
    "\n",
    "Once the environment is setup, you may be required to install some python libraries that are needed to run this notebook. At any point in the notebook, if you receive an error such as: \n",
    "\n",
    " `import 'numpy' is not resolved`\n",
    "\n",
    "Use the pip library to install the library. In case you are running on colab, most of the libraries would already be installed but in case there is a library that isnt on the colab environment use:\n",
    "\n",
    "`!pip install numpy`\n",
    "\n",
    "After installing the Libraries the rest of the cells in this notebook can be run without any issues. \n",
    "\n",
    "--- \n",
    "\n",
    "For the rest of this notebook we will be using popular data science libraries, the links for the documentation would be provided through which any doubt regarding the syntax can be clarified but in case of more doubts please look it up on the internet until you understand what the cell is achieving.\n",
    "\n",
    "Also check out: \n",
    "* [Intro to Numpy](https://www.youtube.com/watch?v=xECXZ3tyONo)\n",
    "\n",
    "* [Intro to Pandas and Matplotlib](https://www.youtube.com/watch?v=a9UrKTVEeZA)\n",
    "\n",
    "---\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='cell2'></a>\n",
    "\n",
    "### Step 2: Importing a Dataset, Some Useful Functions and Dealing with Missing Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Importing the Libraries and a CSV File Dataset into a Pandas DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np \n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Throughout the rest of this notebook we will be adding certain libraries as we require them. \n",
    "* The datasets used for different parts of the notebook may also vary so ensure that you have all the datasets downloaded."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the example here, we are using a dataset that contains Indian Rainfall Data through which we will cover some basic explorations you can"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`pd.read_csv()`\n",
    "\n",
    "This Function allows us to import the comma seperated value file as a Pandas DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('./Datasets/Data_Preprocessing/rainfall.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Some Useful Functions that can be used on DataFrames"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`DataFrame.head()`\n",
    "\n",
    "This Function shows the first 5 entries of the DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>SUBDIVISION</th>\n",
       "      <th>YEAR</th>\n",
       "      <th>JAN</th>\n",
       "      <th>FEB</th>\n",
       "      <th>MAR</th>\n",
       "      <th>APR</th>\n",
       "      <th>MAY</th>\n",
       "      <th>JUN</th>\n",
       "      <th>JUL</th>\n",
       "      <th>AUG</th>\n",
       "      <th>SEP</th>\n",
       "      <th>OCT</th>\n",
       "      <th>NOV</th>\n",
       "      <th>DEC</th>\n",
       "      <th>ANNUAL</th>\n",
       "      <th>Jan-Feb</th>\n",
       "      <th>Mar-May</th>\n",
       "      <th>Jun-Sep</th>\n",
       "      <th>Oct-Dec</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1901</td>\n",
       "      <td>49.2</td>\n",
       "      <td>87.1</td>\n",
       "      <td>29.2</td>\n",
       "      <td>2.3</td>\n",
       "      <td>528.8</td>\n",
       "      <td>517.5</td>\n",
       "      <td>365.1</td>\n",
       "      <td>481.1</td>\n",
       "      <td>332.6</td>\n",
       "      <td>388.5</td>\n",
       "      <td>558.2</td>\n",
       "      <td>33.6</td>\n",
       "      <td>3373.2</td>\n",
       "      <td>136.3</td>\n",
       "      <td>560.3</td>\n",
       "      <td>1696.3</td>\n",
       "      <td>980.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1902</td>\n",
       "      <td>0.0</td>\n",
       "      <td>159.8</td>\n",
       "      <td>12.2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>446.1</td>\n",
       "      <td>537.1</td>\n",
       "      <td>228.9</td>\n",
       "      <td>753.7</td>\n",
       "      <td>666.2</td>\n",
       "      <td>197.2</td>\n",
       "      <td>359.0</td>\n",
       "      <td>160.5</td>\n",
       "      <td>3520.7</td>\n",
       "      <td>159.8</td>\n",
       "      <td>458.3</td>\n",
       "      <td>2185.9</td>\n",
       "      <td>716.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1903</td>\n",
       "      <td>12.7</td>\n",
       "      <td>144.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>235.1</td>\n",
       "      <td>479.9</td>\n",
       "      <td>728.4</td>\n",
       "      <td>326.7</td>\n",
       "      <td>339.0</td>\n",
       "      <td>181.2</td>\n",
       "      <td>284.4</td>\n",
       "      <td>225.0</td>\n",
       "      <td>2957.4</td>\n",
       "      <td>156.7</td>\n",
       "      <td>236.1</td>\n",
       "      <td>1874.0</td>\n",
       "      <td>690.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1904</td>\n",
       "      <td>9.4</td>\n",
       "      <td>14.7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>202.4</td>\n",
       "      <td>304.5</td>\n",
       "      <td>495.1</td>\n",
       "      <td>502.0</td>\n",
       "      <td>160.1</td>\n",
       "      <td>820.4</td>\n",
       "      <td>222.2</td>\n",
       "      <td>308.7</td>\n",
       "      <td>40.1</td>\n",
       "      <td>3079.6</td>\n",
       "      <td>24.1</td>\n",
       "      <td>506.9</td>\n",
       "      <td>1977.6</td>\n",
       "      <td>571.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1905</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>26.9</td>\n",
       "      <td>279.5</td>\n",
       "      <td>628.7</td>\n",
       "      <td>368.7</td>\n",
       "      <td>330.5</td>\n",
       "      <td>297.0</td>\n",
       "      <td>260.7</td>\n",
       "      <td>25.4</td>\n",
       "      <td>344.7</td>\n",
       "      <td>2566.7</td>\n",
       "      <td>1.3</td>\n",
       "      <td>309.7</td>\n",
       "      <td>1624.9</td>\n",
       "      <td>630.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   index                SUBDIVISION  YEAR   JAN    FEB   MAR    APR    MAY  \\\n",
       "0      0  ANDAMAN & NICOBAR ISLANDS  1901  49.2   87.1  29.2    2.3  528.8   \n",
       "1      1  ANDAMAN & NICOBAR ISLANDS  1902   0.0  159.8  12.2    0.0  446.1   \n",
       "2      2  ANDAMAN & NICOBAR ISLANDS  1903  12.7  144.0   0.0    1.0  235.1   \n",
       "3      3  ANDAMAN & NICOBAR ISLANDS  1904   9.4   14.7   0.0  202.4  304.5   \n",
       "4      4  ANDAMAN & NICOBAR ISLANDS  1905   1.3    0.0   3.3   26.9  279.5   \n",
       "\n",
       "     JUN    JUL    AUG    SEP    OCT    NOV    DEC  ANNUAL  Jan-Feb  Mar-May  \\\n",
       "0  517.5  365.1  481.1  332.6  388.5  558.2   33.6  3373.2    136.3    560.3   \n",
       "1  537.1  228.9  753.7  666.2  197.2  359.0  160.5  3520.7    159.8    458.3   \n",
       "2  479.9  728.4  326.7  339.0  181.2  284.4  225.0  2957.4    156.7    236.1   \n",
       "3  495.1  502.0  160.1  820.4  222.2  308.7   40.1  3079.6     24.1    506.9   \n",
       "4  628.7  368.7  330.5  297.0  260.7   25.4  344.7  2566.7      1.3    309.7   \n",
       "\n",
       "   Jun-Sep  Oct-Dec  \n",
       "0   1696.3    980.3  \n",
       "1   2185.9    716.7  \n",
       "2   1874.0    690.6  \n",
       "3   1977.6    571.0  \n",
       "4   1624.9    630.8  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`DataFrame.info()`\n",
    "\n",
    "This Function lets us view some information about the DataFrame Columns and their Datatypes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 4116 entries, 0 to 4115\n",
      "Data columns (total 20 columns):\n",
      " #   Column       Non-Null Count  Dtype  \n",
      "---  ------       --------------  -----  \n",
      " 0   index        4116 non-null   int64  \n",
      " 1   SUBDIVISION  4116 non-null   object \n",
      " 2   YEAR         4116 non-null   int64  \n",
      " 3   JAN          4112 non-null   float64\n",
      " 4   FEB          4113 non-null   float64\n",
      " 5   MAR          4110 non-null   float64\n",
      " 6   APR          4112 non-null   float64\n",
      " 7   MAY          4113 non-null   float64\n",
      " 8   JUN          4111 non-null   float64\n",
      " 9   JUL          4109 non-null   float64\n",
      " 10  AUG          4112 non-null   float64\n",
      " 11  SEP          4110 non-null   float64\n",
      " 12  OCT          4109 non-null   float64\n",
      " 13  NOV          4105 non-null   float64\n",
      " 14  DEC          4106 non-null   float64\n",
      " 15  ANNUAL       4090 non-null   float64\n",
      " 16  Jan-Feb      4110 non-null   float64\n",
      " 17  Mar-May      4107 non-null   float64\n",
      " 18  Jun-Sep      4106 non-null   float64\n",
      " 19  Oct-Dec      4103 non-null   float64\n",
      "dtypes: float64(17), int64(2), object(1)\n",
      "memory usage: 643.3+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`DataFrame.describe()`\n",
    "\n",
    "This Function displays the statistics about the DataFrame's columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>YEAR</th>\n",
       "      <th>JAN</th>\n",
       "      <th>FEB</th>\n",
       "      <th>MAR</th>\n",
       "      <th>APR</th>\n",
       "      <th>MAY</th>\n",
       "      <th>JUN</th>\n",
       "      <th>JUL</th>\n",
       "      <th>AUG</th>\n",
       "      <th>SEP</th>\n",
       "      <th>OCT</th>\n",
       "      <th>NOV</th>\n",
       "      <th>DEC</th>\n",
       "      <th>ANNUAL</th>\n",
       "      <th>Jan-Feb</th>\n",
       "      <th>Mar-May</th>\n",
       "      <th>Jun-Sep</th>\n",
       "      <th>Oct-Dec</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4116.000000</td>\n",
       "      <td>4116.000000</td>\n",
       "      <td>4112.000000</td>\n",
       "      <td>4113.000000</td>\n",
       "      <td>4110.000000</td>\n",
       "      <td>4112.000000</td>\n",
       "      <td>4113.000000</td>\n",
       "      <td>4111.000000</td>\n",
       "      <td>4109.000000</td>\n",
       "      <td>4112.000000</td>\n",
       "      <td>4110.000000</td>\n",
       "      <td>4109.000000</td>\n",
       "      <td>4105.000000</td>\n",
       "      <td>4106.000000</td>\n",
       "      <td>4090.000000</td>\n",
       "      <td>4110.000000</td>\n",
       "      <td>4107.000000</td>\n",
       "      <td>4106.000000</td>\n",
       "      <td>4103.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2057.500000</td>\n",
       "      <td>1958.218659</td>\n",
       "      <td>18.957320</td>\n",
       "      <td>21.805325</td>\n",
       "      <td>27.359197</td>\n",
       "      <td>43.127432</td>\n",
       "      <td>85.745417</td>\n",
       "      <td>230.234444</td>\n",
       "      <td>347.214334</td>\n",
       "      <td>290.263497</td>\n",
       "      <td>197.361922</td>\n",
       "      <td>95.507009</td>\n",
       "      <td>39.866163</td>\n",
       "      <td>18.870580</td>\n",
       "      <td>1411.008900</td>\n",
       "      <td>40.747786</td>\n",
       "      <td>155.901753</td>\n",
       "      <td>1064.724769</td>\n",
       "      <td>154.100487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1188.331183</td>\n",
       "      <td>33.140898</td>\n",
       "      <td>33.585371</td>\n",
       "      <td>35.909488</td>\n",
       "      <td>46.959424</td>\n",
       "      <td>67.831168</td>\n",
       "      <td>123.234904</td>\n",
       "      <td>234.710758</td>\n",
       "      <td>269.539667</td>\n",
       "      <td>188.770477</td>\n",
       "      <td>135.408345</td>\n",
       "      <td>99.519134</td>\n",
       "      <td>68.685410</td>\n",
       "      <td>42.369611</td>\n",
       "      <td>903.846565</td>\n",
       "      <td>59.308277</td>\n",
       "      <td>201.316965</td>\n",
       "      <td>707.741531</td>\n",
       "      <td>166.942660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1901.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>62.300000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>57.400000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1028.750000</td>\n",
       "      <td>1930.000000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>8.600000</td>\n",
       "      <td>70.350000</td>\n",
       "      <td>175.600000</td>\n",
       "      <td>155.975000</td>\n",
       "      <td>100.525000</td>\n",
       "      <td>14.600000</td>\n",
       "      <td>0.700000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>804.500000</td>\n",
       "      <td>4.100000</td>\n",
       "      <td>24.050000</td>\n",
       "      <td>573.850000</td>\n",
       "      <td>34.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2057.500000</td>\n",
       "      <td>1958.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>6.700000</td>\n",
       "      <td>7.800000</td>\n",
       "      <td>15.700000</td>\n",
       "      <td>36.600000</td>\n",
       "      <td>138.700000</td>\n",
       "      <td>284.800000</td>\n",
       "      <td>259.400000</td>\n",
       "      <td>173.900000</td>\n",
       "      <td>65.200000</td>\n",
       "      <td>9.500000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1121.300000</td>\n",
       "      <td>19.200000</td>\n",
       "      <td>74.800000</td>\n",
       "      <td>881.100000</td>\n",
       "      <td>98.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>3086.250000</td>\n",
       "      <td>1987.000000</td>\n",
       "      <td>22.200000</td>\n",
       "      <td>26.800000</td>\n",
       "      <td>31.300000</td>\n",
       "      <td>49.950000</td>\n",
       "      <td>97.200000</td>\n",
       "      <td>305.150000</td>\n",
       "      <td>418.400000</td>\n",
       "      <td>377.800000</td>\n",
       "      <td>265.800000</td>\n",
       "      <td>148.400000</td>\n",
       "      <td>46.100000</td>\n",
       "      <td>17.500000</td>\n",
       "      <td>1644.775000</td>\n",
       "      <td>50.375000</td>\n",
       "      <td>196.950000</td>\n",
       "      <td>1288.175000</td>\n",
       "      <td>213.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>4115.000000</td>\n",
       "      <td>2015.000000</td>\n",
       "      <td>583.700000</td>\n",
       "      <td>403.500000</td>\n",
       "      <td>605.600000</td>\n",
       "      <td>595.100000</td>\n",
       "      <td>1168.600000</td>\n",
       "      <td>1609.900000</td>\n",
       "      <td>2362.800000</td>\n",
       "      <td>1664.600000</td>\n",
       "      <td>1222.000000</td>\n",
       "      <td>948.300000</td>\n",
       "      <td>648.900000</td>\n",
       "      <td>617.500000</td>\n",
       "      <td>6331.100000</td>\n",
       "      <td>699.500000</td>\n",
       "      <td>1745.800000</td>\n",
       "      <td>4536.900000</td>\n",
       "      <td>1252.500000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             index         YEAR          JAN          FEB          MAR  \\\n",
       "count  4116.000000  4116.000000  4112.000000  4113.000000  4110.000000   \n",
       "mean   2057.500000  1958.218659    18.957320    21.805325    27.359197   \n",
       "std    1188.331183    33.140898    33.585371    35.909488    46.959424   \n",
       "min       0.000000  1901.000000     0.000000     0.000000     0.000000   \n",
       "25%    1028.750000  1930.000000     0.600000     0.600000     1.000000   \n",
       "50%    2057.500000  1958.000000     6.000000     6.700000     7.800000   \n",
       "75%    3086.250000  1987.000000    22.200000    26.800000    31.300000   \n",
       "max    4115.000000  2015.000000   583.700000   403.500000   605.600000   \n",
       "\n",
       "               APR          MAY          JUN          JUL          AUG  \\\n",
       "count  4112.000000  4113.000000  4111.000000  4109.000000  4112.000000   \n",
       "mean     43.127432    85.745417   230.234444   347.214334   290.263497   \n",
       "std      67.831168   123.234904   234.710758   269.539667   188.770477   \n",
       "min       0.000000     0.000000     0.400000     0.000000     0.000000   \n",
       "25%       3.000000     8.600000    70.350000   175.600000   155.975000   \n",
       "50%      15.700000    36.600000   138.700000   284.800000   259.400000   \n",
       "75%      49.950000    97.200000   305.150000   418.400000   377.800000   \n",
       "max     595.100000  1168.600000  1609.900000  2362.800000  1664.600000   \n",
       "\n",
       "               SEP          OCT          NOV          DEC       ANNUAL  \\\n",
       "count  4110.000000  4109.000000  4105.000000  4106.000000  4090.000000   \n",
       "mean    197.361922    95.507009    39.866163    18.870580  1411.008900   \n",
       "std     135.408345    99.519134    68.685410    42.369611   903.846565   \n",
       "min       0.100000     0.000000     0.000000     0.000000    62.300000   \n",
       "25%     100.525000    14.600000     0.700000     0.100000   804.500000   \n",
       "50%     173.900000    65.200000     9.500000     3.000000  1121.300000   \n",
       "75%     265.800000   148.400000    46.100000    17.500000  1644.775000   \n",
       "max    1222.000000   948.300000   648.900000   617.500000  6331.100000   \n",
       "\n",
       "           Jan-Feb      Mar-May      Jun-Sep      Oct-Dec  \n",
       "count  4110.000000  4107.000000  4106.000000  4103.000000  \n",
       "mean     40.747786   155.901753  1064.724769   154.100487  \n",
       "std      59.308277   201.316965   707.741531   166.942660  \n",
       "min       0.000000     0.000000    57.400000     0.000000  \n",
       "25%       4.100000    24.050000   573.850000    34.200000  \n",
       "50%      19.200000    74.800000   881.100000    98.200000  \n",
       "75%      50.375000   196.950000  1288.175000   213.500000  \n",
       "max     699.500000  1745.800000  4536.900000  1252.500000  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Dealing with Missing Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Approach 1: Dropping the Rows that contain the missing data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`DataFrame.isnull().sum()`\n",
    "\n",
    "This snippet shows how one can count the number of Null values in a dataset by Column name."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "index           0\n",
       "SUBDIVISION     0\n",
       "YEAR            0\n",
       "JAN             4\n",
       "FEB             3\n",
       "MAR             6\n",
       "APR             4\n",
       "MAY             3\n",
       "JUN             5\n",
       "JUL             7\n",
       "AUG             4\n",
       "SEP             6\n",
       "OCT             7\n",
       "NOV            11\n",
       "DEC            10\n",
       "ANNUAL         26\n",
       "Jan-Feb         6\n",
       "Mar-May         9\n",
       "Jun-Sep        10\n",
       "Oct-Dec        13\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`DataFrame.dropna()` and  `DataFrame.drop_duplicates()`\n",
    "\n",
    "The first function allows us to drop the null values and the second allows us to drop duplicates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "index          0\n",
       "SUBDIVISION    0\n",
       "YEAR           0\n",
       "JAN            0\n",
       "FEB            0\n",
       "MAR            0\n",
       "APR            0\n",
       "MAY            0\n",
       "JUN            0\n",
       "JUL            0\n",
       "AUG            0\n",
       "SEP            0\n",
       "OCT            0\n",
       "NOV            0\n",
       "DEC            0\n",
       "ANNUAL         0\n",
       "Jan-Feb        0\n",
       "Mar-May        0\n",
       "Jun-Sep        0\n",
       "Oct-Dec        0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.dropna()\n",
    "df = df.drop_duplicates()\n",
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Approach 2: Imputing the Data with Statistical Data such as Mean, Median or Mode"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('./Datasets/Data_Preprocessing/rainfall.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "index           0\n",
       "SUBDIVISION     0\n",
       "YEAR            0\n",
       "JAN             4\n",
       "FEB             3\n",
       "MAR             6\n",
       "APR             4\n",
       "MAY             3\n",
       "JUN             5\n",
       "JUL             7\n",
       "AUG             4\n",
       "SEP             6\n",
       "OCT             7\n",
       "NOV            11\n",
       "DEC            10\n",
       "ANNUAL         26\n",
       "Jan-Feb         6\n",
       "Mar-May         9\n",
       "Jun-Sep        10\n",
       "Oct-Dec        13\n",
       "dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`DataFrame.fillna()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "index           0\n",
       "SUBDIVISION     0\n",
       "YEAR            0\n",
       "JAN             4\n",
       "FEB             3\n",
       "MAR             6\n",
       "APR             4\n",
       "MAY             3\n",
       "JUN             5\n",
       "JUL             7\n",
       "AUG             4\n",
       "SEP             6\n",
       "OCT             7\n",
       "NOV            11\n",
       "DEC            10\n",
       "ANNUAL          0\n",
       "Jan-Feb         6\n",
       "Mar-May         9\n",
       "Jun-Sep        10\n",
       "Oct-Dec        13\n",
       "dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['ANNUAL'] = df['ANNUAL'].fillna(df['ANNUAL'].median())\n",
    "\n",
    "# Similarly the null values can be filled with the mean\n",
    "\n",
    "# df = df['ANNUAL'].fillna(df['ANNUAL'].mean())\n",
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 3: Data Preprocessing - Intro"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The next few sections will contain code snippets of how to implement the data preprocessing methods discussed in the class. The following sections may contain some lines where the objective of the lines are unclear, please refer to the official documentations for the library to understand them better.\n",
    "\n",
    "* [Numpy Documentation]('https://numpy.org/doc/')\n",
    "\n",
    "* [Pandas Documentation]('https://pandas.pydata.org/docs/')\n",
    "\n",
    "* [Matplotlib Documentation]('https://matplotlib.org/stable/index.html')\n",
    "\n",
    "* [Scikitlearn Documentation]('https://scikit-learn.org/0.21/documentation.html')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 3.1: Data Preprocessing - Aggregation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Data aggregation is the process of compiling typically (large) amounts of information from a given database and organizing it into a more consumable and comprehensive medium. Data aggregation can be applied at any scale, from pivot tables to data lakes, in order to summarize information and make conclusions based on data-rich findings. Because of the growing accessibility to information and importance of personalization metrics across the enterprise, the application of data aggregation has become extremely relevant.*\n",
    "\n",
    "* Combining two or more attributes (or objects) into a single attribute (or object)\n",
    "* Purpose\n",
    "    * Data reduction\n",
    "        * Reduce the number of attributes or objects\n",
    "    * Change of scale\n",
    "        * Cities aggregated into regions, states, countries, etc\n",
    "    * More “stable” data\n",
    "        * Aggregated data tends to have less variability \n",
    "\n",
    "Also see:\n",
    "\n",
    "* [What is Data Aggregation?]('https://www.pagerduty.com/resources/learn/what-is-data-aggregation/')\n",
    "\n",
    "* [Methods of Data Aggregation]('https://towardsdatascience.com/3-methods-for-aggregating-data-with-python-pandas-14ceb75b6f6e')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>CPI</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1913</td>\n",
       "      <td>1</td>\n",
       "      <td>9.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1913</td>\n",
       "      <td>2</td>\n",
       "      <td>9.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1913</td>\n",
       "      <td>3</td>\n",
       "      <td>9.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1913</td>\n",
       "      <td>4</td>\n",
       "      <td>9.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1913</td>\n",
       "      <td>5</td>\n",
       "      <td>9.7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Year  Month  CPI\n",
       "0  1913      1  9.8\n",
       "1  1913      2  9.8\n",
       "2  1913      3  9.8\n",
       "3  1913      4  9.8\n",
       "4  1913      5  9.7"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_aggregation = pd.read_csv('./Datasets/Data_Preprocessing/US_CPI_Yearly.csv')\n",
    "df_aggregation.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(df_aggregation['Year'],df_aggregation['CPI'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can group the data above by the year and take the average CPI of the entire year"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CPI</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Year</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1913</th>\n",
       "      <td>9.883333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1914</th>\n",
       "      <td>10.016667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1915</th>\n",
       "      <td>10.108333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1916</th>\n",
       "      <td>10.883333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1917</th>\n",
       "      <td>12.825000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017</th>\n",
       "      <td>245.119583</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018</th>\n",
       "      <td>251.106833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019</th>\n",
       "      <td>255.657417</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020</th>\n",
       "      <td>258.811167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021</th>\n",
       "      <td>267.203000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>109 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             CPI\n",
       "Year            \n",
       "1913    9.883333\n",
       "1914   10.016667\n",
       "1915   10.108333\n",
       "1916   10.883333\n",
       "1917   12.825000\n",
       "...          ...\n",
       "2017  245.119583\n",
       "2018  251.106833\n",
       "2019  255.657417\n",
       "2020  258.811167\n",
       "2021  267.203000\n",
       "\n",
       "[109 rows x 1 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_aggregation_grouped = df_aggregation.groupby(['Year']).mean()[['CPI']]\n",
    "df_aggregation_grouped"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(df_aggregation_grouped.index,df_aggregation_grouped['CPI'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 3.2: Data Preprocessing - Sampling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Data sampling is a statistical analysis technique used to select, manipulate and analyze a representative subset of data points to identify patterns and trends in the larger data set being examined. It enables data scientists, predictive modelers and other data analysts to work with a small, manageable amount of data about a statistical population to build and run analytical models more quickly, while still producing accurate findings.*\n",
    "\n",
    "* The key principle for effective sampling is: **A sample will work almost as well as using the entire data set if the sample is representative(different for different data set).**\n",
    "* Sampling may remove outliers and if done improperly can introduce noise.\n",
    "\n",
    "##### Types of Sampling:\n",
    "* Simple Random Sampling\n",
    "    * There is an equal probability of selecting any particular item\n",
    "* Sampling without replacement\n",
    "    * As each item is selected, it is removed from the population\n",
    "* Sampling with replacement\n",
    "    * Objects are not removed from the population as they are selected for the sample.   \n",
    "        * In sampling with replacement, the same object can be picked up more than once\n",
    "* Stratified sampling\n",
    "    * Split the data into several partitions; then draw random samples from each partition\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>SUBDIVISION</th>\n",
       "      <th>YEAR</th>\n",
       "      <th>JAN</th>\n",
       "      <th>FEB</th>\n",
       "      <th>MAR</th>\n",
       "      <th>APR</th>\n",
       "      <th>MAY</th>\n",
       "      <th>JUN</th>\n",
       "      <th>JUL</th>\n",
       "      <th>AUG</th>\n",
       "      <th>SEP</th>\n",
       "      <th>OCT</th>\n",
       "      <th>NOV</th>\n",
       "      <th>DEC</th>\n",
       "      <th>ANNUAL</th>\n",
       "      <th>Jan-Feb</th>\n",
       "      <th>Mar-May</th>\n",
       "      <th>Jun-Sep</th>\n",
       "      <th>Oct-Dec</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1901</td>\n",
       "      <td>49.2</td>\n",
       "      <td>87.1</td>\n",
       "      <td>29.2</td>\n",
       "      <td>2.3</td>\n",
       "      <td>528.8</td>\n",
       "      <td>517.5</td>\n",
       "      <td>365.1</td>\n",
       "      <td>481.1</td>\n",
       "      <td>332.6</td>\n",
       "      <td>388.5</td>\n",
       "      <td>558.2</td>\n",
       "      <td>33.6</td>\n",
       "      <td>3373.2</td>\n",
       "      <td>136.3</td>\n",
       "      <td>560.3</td>\n",
       "      <td>1696.3</td>\n",
       "      <td>980.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1902</td>\n",
       "      <td>0.0</td>\n",
       "      <td>159.8</td>\n",
       "      <td>12.2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>446.1</td>\n",
       "      <td>537.1</td>\n",
       "      <td>228.9</td>\n",
       "      <td>753.7</td>\n",
       "      <td>666.2</td>\n",
       "      <td>197.2</td>\n",
       "      <td>359.0</td>\n",
       "      <td>160.5</td>\n",
       "      <td>3520.7</td>\n",
       "      <td>159.8</td>\n",
       "      <td>458.3</td>\n",
       "      <td>2185.9</td>\n",
       "      <td>716.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1903</td>\n",
       "      <td>12.7</td>\n",
       "      <td>144.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>235.1</td>\n",
       "      <td>479.9</td>\n",
       "      <td>728.4</td>\n",
       "      <td>326.7</td>\n",
       "      <td>339.0</td>\n",
       "      <td>181.2</td>\n",
       "      <td>284.4</td>\n",
       "      <td>225.0</td>\n",
       "      <td>2957.4</td>\n",
       "      <td>156.7</td>\n",
       "      <td>236.1</td>\n",
       "      <td>1874.0</td>\n",
       "      <td>690.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1904</td>\n",
       "      <td>9.4</td>\n",
       "      <td>14.7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>202.4</td>\n",
       "      <td>304.5</td>\n",
       "      <td>495.1</td>\n",
       "      <td>502.0</td>\n",
       "      <td>160.1</td>\n",
       "      <td>820.4</td>\n",
       "      <td>222.2</td>\n",
       "      <td>308.7</td>\n",
       "      <td>40.1</td>\n",
       "      <td>3079.6</td>\n",
       "      <td>24.1</td>\n",
       "      <td>506.9</td>\n",
       "      <td>1977.6</td>\n",
       "      <td>571.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1905</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>26.9</td>\n",
       "      <td>279.5</td>\n",
       "      <td>628.7</td>\n",
       "      <td>368.7</td>\n",
       "      <td>330.5</td>\n",
       "      <td>297.0</td>\n",
       "      <td>260.7</td>\n",
       "      <td>25.4</td>\n",
       "      <td>344.7</td>\n",
       "      <td>2566.7</td>\n",
       "      <td>1.3</td>\n",
       "      <td>309.7</td>\n",
       "      <td>1624.9</td>\n",
       "      <td>630.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   index                SUBDIVISION  YEAR   JAN    FEB   MAR    APR    MAY  \\\n",
       "0      0  ANDAMAN & NICOBAR ISLANDS  1901  49.2   87.1  29.2    2.3  528.8   \n",
       "1      1  ANDAMAN & NICOBAR ISLANDS  1902   0.0  159.8  12.2    0.0  446.1   \n",
       "2      2  ANDAMAN & NICOBAR ISLANDS  1903  12.7  144.0   0.0    1.0  235.1   \n",
       "3      3  ANDAMAN & NICOBAR ISLANDS  1904   9.4   14.7   0.0  202.4  304.5   \n",
       "4      4  ANDAMAN & NICOBAR ISLANDS  1905   1.3    0.0   3.3   26.9  279.5   \n",
       "\n",
       "     JUN    JUL    AUG    SEP    OCT    NOV    DEC  ANNUAL  Jan-Feb  Mar-May  \\\n",
       "0  517.5  365.1  481.1  332.6  388.5  558.2   33.6  3373.2    136.3    560.3   \n",
       "1  537.1  228.9  753.7  666.2  197.2  359.0  160.5  3520.7    159.8    458.3   \n",
       "2  479.9  728.4  326.7  339.0  181.2  284.4  225.0  2957.4    156.7    236.1   \n",
       "3  495.1  502.0  160.1  820.4  222.2  308.7   40.1  3079.6     24.1    506.9   \n",
       "4  628.7  368.7  330.5  297.0  260.7   25.4  344.7  2566.7      1.3    309.7   \n",
       "\n",
       "   Jun-Sep  Oct-Dec  \n",
       "0   1696.3    980.3  \n",
       "1   2185.9    716.7  \n",
       "2   1874.0    690.6  \n",
       "3   1977.6    571.0  \n",
       "4   1624.9    630.8  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sampling = pd.read_csv('./Datasets/Data_Preprocessing/rainfall.csv')\n",
    "df_sampling.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Simple Random Sampling without Replacement"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`DataFrame.sample()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>SUBDIVISION</th>\n",
       "      <th>YEAR</th>\n",
       "      <th>JAN</th>\n",
       "      <th>FEB</th>\n",
       "      <th>MAR</th>\n",
       "      <th>APR</th>\n",
       "      <th>MAY</th>\n",
       "      <th>JUN</th>\n",
       "      <th>JUL</th>\n",
       "      <th>AUG</th>\n",
       "      <th>SEP</th>\n",
       "      <th>OCT</th>\n",
       "      <th>NOV</th>\n",
       "      <th>DEC</th>\n",
       "      <th>ANNUAL</th>\n",
       "      <th>Jan-Feb</th>\n",
       "      <th>Mar-May</th>\n",
       "      <th>Jun-Sep</th>\n",
       "      <th>Oct-Dec</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>425</th>\n",
       "      <td>425</td>\n",
       "      <td>NAGA MANI MIZO TRIPURA</td>\n",
       "      <td>2004</td>\n",
       "      <td>4.8</td>\n",
       "      <td>2.7</td>\n",
       "      <td>11.0</td>\n",
       "      <td>341.5</td>\n",
       "      <td>211.0</td>\n",
       "      <td>449.2</td>\n",
       "      <td>558.6</td>\n",
       "      <td>284.3</td>\n",
       "      <td>351.0</td>\n",
       "      <td>132.7</td>\n",
       "      <td>6.4</td>\n",
       "      <td>0.5</td>\n",
       "      <td>2353.7</td>\n",
       "      <td>7.4</td>\n",
       "      <td>563.5</td>\n",
       "      <td>1643.2</td>\n",
       "      <td>139.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2074</th>\n",
       "      <td>2074</td>\n",
       "      <td>WEST MADHYA PRADESH</td>\n",
       "      <td>1928</td>\n",
       "      <td>0.4</td>\n",
       "      <td>31.9</td>\n",
       "      <td>2.8</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>75.6</td>\n",
       "      <td>395.0</td>\n",
       "      <td>217.6</td>\n",
       "      <td>49.5</td>\n",
       "      <td>66.9</td>\n",
       "      <td>1.6</td>\n",
       "      <td>51.9</td>\n",
       "      <td>895.6</td>\n",
       "      <td>32.3</td>\n",
       "      <td>5.3</td>\n",
       "      <td>737.7</td>\n",
       "      <td>120.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3658</th>\n",
       "      <td>3658</td>\n",
       "      <td>NORTH INTERIOR KARNATAKA</td>\n",
       "      <td>1902</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>22.5</td>\n",
       "      <td>34.4</td>\n",
       "      <td>111.3</td>\n",
       "      <td>83.2</td>\n",
       "      <td>78.1</td>\n",
       "      <td>146.7</td>\n",
       "      <td>118.8</td>\n",
       "      <td>35.7</td>\n",
       "      <td>85.1</td>\n",
       "      <td>716.1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57.1</td>\n",
       "      <td>419.3</td>\n",
       "      <td>239.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3567</th>\n",
       "      <td>3567</td>\n",
       "      <td>COASTAL KARNATAKA</td>\n",
       "      <td>1926</td>\n",
       "      <td>12.3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.1</td>\n",
       "      <td>7.3</td>\n",
       "      <td>15.3</td>\n",
       "      <td>686.3</td>\n",
       "      <td>976.4</td>\n",
       "      <td>850.3</td>\n",
       "      <td>258.7</td>\n",
       "      <td>143.6</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.1</td>\n",
       "      <td>2956.4</td>\n",
       "      <td>14.3</td>\n",
       "      <td>22.6</td>\n",
       "      <td>2771.7</td>\n",
       "      <td>147.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3774</th>\n",
       "      <td>3774</td>\n",
       "      <td>SOUTH INTERIOR KARNATAKA</td>\n",
       "      <td>1903</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.1</td>\n",
       "      <td>11.6</td>\n",
       "      <td>125.1</td>\n",
       "      <td>129.7</td>\n",
       "      <td>284.4</td>\n",
       "      <td>155.7</td>\n",
       "      <td>197.1</td>\n",
       "      <td>154.2</td>\n",
       "      <td>186.6</td>\n",
       "      <td>24.1</td>\n",
       "      <td>1269.9</td>\n",
       "      <td>0.3</td>\n",
       "      <td>137.7</td>\n",
       "      <td>766.9</td>\n",
       "      <td>364.9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      index               SUBDIVISION  YEAR   JAN   FEB   MAR    APR    MAY  \\\n",
       "425     425    NAGA MANI MIZO TRIPURA  2004   4.8   2.7  11.0  341.5  211.0   \n",
       "2074   2074       WEST MADHYA PRADESH  1928   0.4  31.9   2.8    0.0    2.5   \n",
       "3658   3658  NORTH INTERIOR KARNATAKA  1902   0.0   0.0   0.3   22.5   34.4   \n",
       "3567   3567         COASTAL KARNATAKA  1926  12.3   2.0   0.1    7.3   15.3   \n",
       "3774   3774  SOUTH INTERIOR KARNATAKA  1903   0.3   0.0   1.1   11.6  125.1   \n",
       "\n",
       "        JUN    JUL    AUG    SEP    OCT    NOV   DEC  ANNUAL  Jan-Feb  \\\n",
       "425   449.2  558.6  284.3  351.0  132.7    6.4   0.5  2353.7      7.4   \n",
       "2074   75.6  395.0  217.6   49.5   66.9    1.6  51.9   895.6     32.3   \n",
       "3658  111.3   83.2   78.1  146.7  118.8   35.7  85.1   716.1      0.0   \n",
       "3567  686.3  976.4  850.3  258.7  143.6    4.0   0.1  2956.4     14.3   \n",
       "3774  129.7  284.4  155.7  197.1  154.2  186.6  24.1  1269.9      0.3   \n",
       "\n",
       "      Mar-May  Jun-Sep  Oct-Dec  \n",
       "425     563.5   1643.2    139.6  \n",
       "2074      5.3    737.7    120.3  \n",
       "3658     57.1    419.3    239.6  \n",
       "3567     22.6   2771.7    147.7  \n",
       "3774    137.7    766.9    364.9  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "random_sampled_df = df_sampling.sample(36) # Sampling 10 random items\n",
    "random_sampled_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Simple Random Sampling with Replacement"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`DataFrame.sample(replace=True)`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>SUBDIVISION</th>\n",
       "      <th>YEAR</th>\n",
       "      <th>JAN</th>\n",
       "      <th>FEB</th>\n",
       "      <th>MAR</th>\n",
       "      <th>APR</th>\n",
       "      <th>MAY</th>\n",
       "      <th>JUN</th>\n",
       "      <th>JUL</th>\n",
       "      <th>AUG</th>\n",
       "      <th>SEP</th>\n",
       "      <th>OCT</th>\n",
       "      <th>NOV</th>\n",
       "      <th>DEC</th>\n",
       "      <th>ANNUAL</th>\n",
       "      <th>Jan-Feb</th>\n",
       "      <th>Mar-May</th>\n",
       "      <th>Jun-Sep</th>\n",
       "      <th>Oct-Dec</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>873</th>\n",
       "      <td>873</td>\n",
       "      <td>JHARKHAND</td>\n",
       "      <td>1992</td>\n",
       "      <td>2.3</td>\n",
       "      <td>6.1</td>\n",
       "      <td>0.6</td>\n",
       "      <td>8.8</td>\n",
       "      <td>51.8</td>\n",
       "      <td>129.8</td>\n",
       "      <td>295.7</td>\n",
       "      <td>265.2</td>\n",
       "      <td>148.5</td>\n",
       "      <td>22.5</td>\n",
       "      <td>1.6</td>\n",
       "      <td>0.3</td>\n",
       "      <td>933.0</td>\n",
       "      <td>8.3</td>\n",
       "      <td>61.2</td>\n",
       "      <td>839.2</td>\n",
       "      <td>24.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1508</th>\n",
       "      <td>1508</td>\n",
       "      <td>PUNJAB</td>\n",
       "      <td>1937</td>\n",
       "      <td>5.8</td>\n",
       "      <td>96.0</td>\n",
       "      <td>10.2</td>\n",
       "      <td>41.8</td>\n",
       "      <td>8.8</td>\n",
       "      <td>55.4</td>\n",
       "      <td>218.6</td>\n",
       "      <td>30.5</td>\n",
       "      <td>70.2</td>\n",
       "      <td>7.3</td>\n",
       "      <td>0.3</td>\n",
       "      <td>26.2</td>\n",
       "      <td>571.1</td>\n",
       "      <td>101.8</td>\n",
       "      <td>60.8</td>\n",
       "      <td>374.7</td>\n",
       "      <td>33.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3490</th>\n",
       "      <td>3490</td>\n",
       "      <td>TAMIL NADU</td>\n",
       "      <td>1964</td>\n",
       "      <td>1.6</td>\n",
       "      <td>1.2</td>\n",
       "      <td>15.5</td>\n",
       "      <td>11.4</td>\n",
       "      <td>56.8</td>\n",
       "      <td>31.3</td>\n",
       "      <td>174.7</td>\n",
       "      <td>100.8</td>\n",
       "      <td>92.3</td>\n",
       "      <td>187.7</td>\n",
       "      <td>166.4</td>\n",
       "      <td>70.9</td>\n",
       "      <td>910.6</td>\n",
       "      <td>2.7</td>\n",
       "      <td>83.8</td>\n",
       "      <td>399.1</td>\n",
       "      <td>424.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>19</td>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1921</td>\n",
       "      <td>13.2</td>\n",
       "      <td>3.1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>37.5</td>\n",
       "      <td>351.2</td>\n",
       "      <td>282.7</td>\n",
       "      <td>487.1</td>\n",
       "      <td>330.0</td>\n",
       "      <td>581.2</td>\n",
       "      <td>360.7</td>\n",
       "      <td>118.2</td>\n",
       "      <td>41.5</td>\n",
       "      <td>2606.4</td>\n",
       "      <td>16.3</td>\n",
       "      <td>388.7</td>\n",
       "      <td>1681.0</td>\n",
       "      <td>520.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1494</th>\n",
       "      <td>1494</td>\n",
       "      <td>PUNJAB</td>\n",
       "      <td>1923</td>\n",
       "      <td>36.4</td>\n",
       "      <td>63.0</td>\n",
       "      <td>7.3</td>\n",
       "      <td>4.9</td>\n",
       "      <td>38.0</td>\n",
       "      <td>21.8</td>\n",
       "      <td>222.2</td>\n",
       "      <td>279.1</td>\n",
       "      <td>12.8</td>\n",
       "      <td>5.7</td>\n",
       "      <td>2.5</td>\n",
       "      <td>44.1</td>\n",
       "      <td>737.9</td>\n",
       "      <td>99.4</td>\n",
       "      <td>50.2</td>\n",
       "      <td>536.0</td>\n",
       "      <td>52.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      index                SUBDIVISION  YEAR   JAN   FEB   MAR   APR    MAY  \\\n",
       "873     873                  JHARKHAND  1992   2.3   6.1   0.6   8.8   51.8   \n",
       "1508   1508                     PUNJAB  1937   5.8  96.0  10.2  41.8    8.8   \n",
       "3490   3490                 TAMIL NADU  1964   1.6   1.2  15.5  11.4   56.8   \n",
       "19       19  ANDAMAN & NICOBAR ISLANDS  1921  13.2   3.1   0.0  37.5  351.2   \n",
       "1494   1494                     PUNJAB  1923  36.4  63.0   7.3   4.9   38.0   \n",
       "\n",
       "        JUN    JUL    AUG    SEP    OCT    NOV   DEC  ANNUAL  Jan-Feb  \\\n",
       "873   129.8  295.7  265.2  148.5   22.5    1.6   0.3   933.0      8.3   \n",
       "1508   55.4  218.6   30.5   70.2    7.3    0.3  26.2   571.1    101.8   \n",
       "3490   31.3  174.7  100.8   92.3  187.7  166.4  70.9   910.6      2.7   \n",
       "19    282.7  487.1  330.0  581.2  360.7  118.2  41.5  2606.4     16.3   \n",
       "1494   21.8  222.2  279.1   12.8    5.7    2.5  44.1   737.9     99.4   \n",
       "\n",
       "      Mar-May  Jun-Sep  Oct-Dec  \n",
       "873      61.2    839.2     24.3  \n",
       "1508     60.8    374.7     33.8  \n",
       "3490     83.8    399.1    424.9  \n",
       "19      388.7   1681.0    520.4  \n",
       "1494     50.2    536.0     52.3  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "random_sampled_replacement_df = df_sampling.sample(36,replace=True) # Sampling 10 random items\n",
    "random_sampled_replacement_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Stratified Sampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_20425/1616637742.py:3: DeprecationWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning.\n",
      "  stratified_sampled_df = df_stratified_sampling.groupby('SUBDIVISION',group_keys=False).apply(lambda x: x.sample(frac=0.01))\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SUBDIVISION</th>\n",
       "      <th>YEAR</th>\n",
       "      <th>JAN</th>\n",
       "      <th>FEB</th>\n",
       "      <th>MAR</th>\n",
       "      <th>APR</th>\n",
       "      <th>MAY</th>\n",
       "      <th>JUN</th>\n",
       "      <th>JUL</th>\n",
       "      <th>AUG</th>\n",
       "      <th>SEP</th>\n",
       "      <th>OCT</th>\n",
       "      <th>NOV</th>\n",
       "      <th>DEC</th>\n",
       "      <th>ANNUAL</th>\n",
       "      <th>Jan-Feb</th>\n",
       "      <th>Mar-May</th>\n",
       "      <th>Jun-Sep</th>\n",
       "      <th>Oct-Dec</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ANDAMAN &amp; NICOBAR ISLANDS</td>\n",
       "      <td>1901</td>\n",
       "      <td>49.2</td>\n",
       "      <td>87.1</td>\n",
       "      <td>29.2</td>\n",
       "      <td>2.3</td>\n",
       "      <td>528.8</td>\n",
       "      <td>517.5</td>\n",
       "      <td>365.1</td>\n",
       "      <td>481.1</td>\n",
       "      <td>332.6</td>\n",
       "      <td>388.5</td>\n",
       "      <td>558.2</td>\n",
       "      <td>33.6</td>\n",
       "      <td>3373.2</td>\n",
       "      <td>136.3</td>\n",
       "      <td>560.3</td>\n",
       "      <td>1696.3</td>\n",
       "      <td>980.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>ARUNACHAL PRADESH</td>\n",
       "      <td>1930</td>\n",
       "      <td>64.4</td>\n",
       "      <td>84.3</td>\n",
       "      <td>202.5</td>\n",
       "      <td>565.4</td>\n",
       "      <td>148.9</td>\n",
       "      <td>720.9</td>\n",
       "      <td>376.2</td>\n",
       "      <td>1055.5</td>\n",
       "      <td>637.0</td>\n",
       "      <td>361.1</td>\n",
       "      <td>93.2</td>\n",
       "      <td>24.9</td>\n",
       "      <td>4334.1</td>\n",
       "      <td>148.7</td>\n",
       "      <td>916.8</td>\n",
       "      <td>2789.5</td>\n",
       "      <td>479.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>252</th>\n",
       "      <td>ASSAM &amp; MEGHALAYA</td>\n",
       "      <td>1946</td>\n",
       "      <td>0.4</td>\n",
       "      <td>19.4</td>\n",
       "      <td>102.6</td>\n",
       "      <td>178.5</td>\n",
       "      <td>328.0</td>\n",
       "      <td>582.6</td>\n",
       "      <td>547.0</td>\n",
       "      <td>297.9</td>\n",
       "      <td>266.8</td>\n",
       "      <td>346.3</td>\n",
       "      <td>8.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>2679.5</td>\n",
       "      <td>19.8</td>\n",
       "      <td>609.2</td>\n",
       "      <td>1694.2</td>\n",
       "      <td>356.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>951</th>\n",
       "      <td>BIHAR</td>\n",
       "      <td>1955</td>\n",
       "      <td>20.0</td>\n",
       "      <td>6.7</td>\n",
       "      <td>2.3</td>\n",
       "      <td>11.9</td>\n",
       "      <td>28.2</td>\n",
       "      <td>172.5</td>\n",
       "      <td>505.7</td>\n",
       "      <td>268.7</td>\n",
       "      <td>192.2</td>\n",
       "      <td>35.7</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1244.2</td>\n",
       "      <td>26.7</td>\n",
       "      <td>42.5</td>\n",
       "      <td>1139.0</td>\n",
       "      <td>36.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3069</th>\n",
       "      <td>CHHATTISGARH</td>\n",
       "      <td>2003</td>\n",
       "      <td>3.7</td>\n",
       "      <td>17.7</td>\n",
       "      <td>12.1</td>\n",
       "      <td>2.6</td>\n",
       "      <td>0.7</td>\n",
       "      <td>128.7</td>\n",
       "      <td>446.6</td>\n",
       "      <td>519.7</td>\n",
       "      <td>384.1</td>\n",
       "      <td>157.6</td>\n",
       "      <td>5.8</td>\n",
       "      <td>14.9</td>\n",
       "      <td>1694.3</td>\n",
       "      <td>21.4</td>\n",
       "      <td>15.5</td>\n",
       "      <td>1479.1</td>\n",
       "      <td>178.2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    SUBDIVISION  YEAR   JAN   FEB    MAR    APR    MAY    JUN  \\\n",
       "0     ANDAMAN & NICOBAR ISLANDS  1901  49.2  87.1   29.2    2.3  528.8  517.5   \n",
       "124           ARUNACHAL PRADESH  1930  64.4  84.3  202.5  565.4  148.9  720.9   \n",
       "252           ASSAM & MEGHALAYA  1946   0.4  19.4  102.6  178.5  328.0  582.6   \n",
       "951                       BIHAR  1955  20.0   6.7    2.3   11.9   28.2  172.5   \n",
       "3069               CHHATTISGARH  2003   3.7  17.7   12.1    2.6    0.7  128.7   \n",
       "\n",
       "        JUL     AUG    SEP    OCT    NOV   DEC  ANNUAL  Jan-Feb  Mar-May  \\\n",
       "0     365.1   481.1  332.6  388.5  558.2  33.6  3373.2    136.3    560.3   \n",
       "124   376.2  1055.5  637.0  361.1   93.2  24.9  4334.1    148.7    916.8   \n",
       "252   547.0   297.9  266.8  346.3    8.6   1.4  2679.5     19.8    609.2   \n",
       "951   505.7   268.7  192.2   35.7    0.3   0.1  1244.2     26.7     42.5   \n",
       "3069  446.6   519.7  384.1  157.6    5.8  14.9  1694.3     21.4     15.5   \n",
       "\n",
       "      Jun-Sep  Oct-Dec  \n",
       "0      1696.3    980.3  \n",
       "124    2789.5    479.1  \n",
       "252    1694.2    356.4  \n",
       "951    1139.0     36.1  \n",
       "3069   1479.1    178.2  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stratified_sampling = pd.read_csv('./Datasets/Data_Preprocessing/rainfall.csv')\n",
    "df_stratified_sampling = df_stratified_sampling.drop('index',axis=1)\n",
    "stratified_sampled_df = df_stratified_sampling.groupby('SUBDIVISION',group_keys=False).apply(lambda x: x.sample(frac=0.01))\n",
    "stratified_sampled_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The total number of Unique Subdivisions in the Dataset: 36\n"
     ]
    }
   ],
   "source": [
    "print(f\"The total number of Unique Subdivisions in the Dataset: {len(df_stratified_sampling['SUBDIVISION'].unique())}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total Number of Items Sampled: 36\n"
     ]
    }
   ],
   "source": [
    "print(f\"Total Number of Items Sampled: {stratified_sampled_df['SUBDIVISION'].value_counts().sum()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Uncomment this to verify that there is infact 1 sample from every subdivision"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# stratified_sampled_df['SUBDIVISION'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that the data is sampled equally from each of the states as compared to the previous random samples as compared to the randomly sampled instances above. This is the advantage that stratified sampling provides.\n",
    "\n",
    "Also see:\n",
    "* [More Sampling Methods and their Implementations]('https://medium.com/lunartechai/data-sampling-methods-in-python-a4400628ea1b')\n",
    "* [Stratified Sampling v/s Random Sampling]('https://www.investopedia.com/ask/answers/042415/what-difference-between-simple-random-sample-and-stratified-random-sample.asp')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 3.3 Data Preprocessing - Dimensionality Reduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### **Dimensionality Reduction using Principal Component Analysis**\n",
    "\n",
    "*Dimensionality reduction, or dimension reduction, is the transformation of data from a high-dimensional space into a low-dimensional space so that the low-dimensional representation retains some meaningful properties of the original data, ideally close to its intrinsic dimension.Principal component analysis, or PCA, is a dimensionality reduction method that is often used to reduce the dimensionality of large data sets, by transforming a large set of variables into a smaller one that still contains most of the information in the large set*\n",
    "\n",
    "---\n",
    "\n",
    "##### **Working of PCA**:\n",
    "\n",
    "**1. Mean Center the Data**\n",
    "    \n",
    "The mean of the data can be computed by using the formula:\n",
    "$$ \\bar{x} = \\frac{\\sum_{n=1}^N x_n}{N} $$\n",
    "\n",
    "Then to Mean Center the data simply replace every data point $x_i$ as follows: $$x_i = x_i - \\bar{x}$$ \n",
    "\n",
    "When the data is Mean Centered, the new mean of the data $\\bar{x}_{new}$ becomes: \n",
    "$$ \\bar{x}_{new} = 0 $$\n",
    "\n",
    "**2. Compute the Covariance Matrix**\n",
    "\n",
    "The covariance between x and y can be computed using the formula:\n",
    "$$ cov(x,y) = \\frac{\\sum_{i=1^n}(x_i - \\bar{x})(y_i - \\bar{y})}{n-1} $$\n",
    "\n",
    "**3. Compute the Eigen Value Decomposition of the Covariance Matrix**\n",
    "\n",
    "Let $X$ be a matrix such that:\n",
    "$$\n",
    "X =\n",
    "\\begin{pmatrix}\n",
    "  x_{n1}   \\\\\n",
    "  x_{n2}  \\\\\n",
    "  \\vdots  \\\\\n",
    "  x_{nD}   \\\\\n",
    "\\end{pmatrix}\n",
    "$$\n",
    "\n",
    "Let $\\bar{X}$ be a matrix such that:\n",
    "$$\n",
    "\\bar{X} = \n",
    "\\begin{pmatrix}\n",
    "  \\bar{x}_1 & \\bar{x}_2 & \\dots & \\bar{x}_n   \\\\\n",
    "\\end{pmatrix}\n",
    "$$\n",
    "\n",
    "The Covariance Matrix $S$ is given by:\n",
    "\n",
    "$$ S = \\frac{\\sum_{n=1}^N (X_n - \\bar{X}) (X_n - \\bar{X})^T}{N} $$\n",
    "\n",
    "Once this matrix $S$ is derived, compute its Eigenvalues\n",
    "\n",
    "**4. Find the Best $k$ Eigen Vectors by using Eigen Values**\n",
    "\n",
    "The dimensions of your dataset are now reduced to $k$ dimensions instead of $N$\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Alcohol</th>\n",
       "      <th>Malic_Acid</th>\n",
       "      <th>Ash</th>\n",
       "      <th>Ash_Alcanity</th>\n",
       "      <th>Magnesium</th>\n",
       "      <th>Total_Phenols</th>\n",
       "      <th>Flavanoids</th>\n",
       "      <th>Nonflavanoid_Phenols</th>\n",
       "      <th>Proanthocyanins</th>\n",
       "      <th>Color_Intensity</th>\n",
       "      <th>Hue</th>\n",
       "      <th>OD280</th>\n",
       "      <th>Proline</th>\n",
       "      <th>Customer_Segment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>14.23</td>\n",
       "      <td>1.71</td>\n",
       "      <td>2.43</td>\n",
       "      <td>15.6</td>\n",
       "      <td>127</td>\n",
       "      <td>2.80</td>\n",
       "      <td>3.06</td>\n",
       "      <td>0.28</td>\n",
       "      <td>2.29</td>\n",
       "      <td>5.64</td>\n",
       "      <td>1.04</td>\n",
       "      <td>3.92</td>\n",
       "      <td>1065</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>13.20</td>\n",
       "      <td>1.78</td>\n",
       "      <td>2.14</td>\n",
       "      <td>11.2</td>\n",
       "      <td>100</td>\n",
       "      <td>2.65</td>\n",
       "      <td>2.76</td>\n",
       "      <td>0.26</td>\n",
       "      <td>1.28</td>\n",
       "      <td>4.38</td>\n",
       "      <td>1.05</td>\n",
       "      <td>3.40</td>\n",
       "      <td>1050</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>13.16</td>\n",
       "      <td>2.36</td>\n",
       "      <td>2.67</td>\n",
       "      <td>18.6</td>\n",
       "      <td>101</td>\n",
       "      <td>2.80</td>\n",
       "      <td>3.24</td>\n",
       "      <td>0.30</td>\n",
       "      <td>2.81</td>\n",
       "      <td>5.68</td>\n",
       "      <td>1.03</td>\n",
       "      <td>3.17</td>\n",
       "      <td>1185</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>14.37</td>\n",
       "      <td>1.95</td>\n",
       "      <td>2.50</td>\n",
       "      <td>16.8</td>\n",
       "      <td>113</td>\n",
       "      <td>3.85</td>\n",
       "      <td>3.49</td>\n",
       "      <td>0.24</td>\n",
       "      <td>2.18</td>\n",
       "      <td>7.80</td>\n",
       "      <td>0.86</td>\n",
       "      <td>3.45</td>\n",
       "      <td>1480</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>13.24</td>\n",
       "      <td>2.59</td>\n",
       "      <td>2.87</td>\n",
       "      <td>21.0</td>\n",
       "      <td>118</td>\n",
       "      <td>2.80</td>\n",
       "      <td>2.69</td>\n",
       "      <td>0.39</td>\n",
       "      <td>1.82</td>\n",
       "      <td>4.32</td>\n",
       "      <td>1.04</td>\n",
       "      <td>2.93</td>\n",
       "      <td>735</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Alcohol  Malic_Acid   Ash  Ash_Alcanity  Magnesium  Total_Phenols  \\\n",
       "0    14.23        1.71  2.43          15.6        127           2.80   \n",
       "1    13.20        1.78  2.14          11.2        100           2.65   \n",
       "2    13.16        2.36  2.67          18.6        101           2.80   \n",
       "3    14.37        1.95  2.50          16.8        113           3.85   \n",
       "4    13.24        2.59  2.87          21.0        118           2.80   \n",
       "\n",
       "   Flavanoids  Nonflavanoid_Phenols  Proanthocyanins  Color_Intensity   Hue  \\\n",
       "0        3.06                  0.28             2.29             5.64  1.04   \n",
       "1        2.76                  0.26             1.28             4.38  1.05   \n",
       "2        3.24                  0.30             2.81             5.68  1.03   \n",
       "3        3.49                  0.24             2.18             7.80  0.86   \n",
       "4        2.69                  0.39             1.82             4.32  1.04   \n",
       "\n",
       "   OD280  Proline  Customer_Segment  \n",
       "0   3.92     1065                 1  \n",
       "1   3.40     1050                 1  \n",
       "2   3.17     1185                 1  \n",
       "3   3.45     1480                 1  \n",
       "4   2.93      735                 1  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca_df = pd.read_csv('./Datasets/Data_Preprocessing/Wine.csv')\n",
    "pca_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Number of Dimensions in the Dataset: 13\n"
     ]
    }
   ],
   "source": [
    "print(f\"The Number of Dimensions in the Dataset: {len(pca_df.columns[:-1])}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.423e+01, 1.710e+00, 2.430e+00, ..., 1.040e+00, 3.920e+00,\n",
       "        1.065e+03],\n",
       "       [1.320e+01, 1.780e+00, 2.140e+00, ..., 1.050e+00, 3.400e+00,\n",
       "        1.050e+03],\n",
       "       [1.316e+01, 2.360e+00, 2.670e+00, ..., 1.030e+00, 3.170e+00,\n",
       "        1.185e+03],\n",
       "       ...,\n",
       "       [1.327e+01, 4.280e+00, 2.260e+00, ..., 5.900e-01, 1.560e+00,\n",
       "        8.350e+02],\n",
       "       [1.317e+01, 2.590e+00, 2.370e+00, ..., 6.000e-01, 1.620e+00,\n",
       "        8.400e+02],\n",
       "       [1.413e+01, 4.100e+00, 2.740e+00, ..., 6.100e-01, 1.600e+00,\n",
       "        5.600e+02]])"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = pca_df.iloc[:, 0:13].values\n",
    "y = pca_df.iloc[:, 13].values\n",
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.decomposition import PCA\n",
    " \n",
    "pca = PCA(n_components = 3)\n",
    " \n",
    "X_train = pca.fit_transform(X)\n",
    "X_test = pca.transform(X)\n",
    " \n",
    "explained_variance = pca.explained_variance_ratio_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The top 3 Variance Ratios are: [9.98091230e-01 1.73591562e-03 9.49589576e-05]\n"
     ]
    }
   ],
   "source": [
    "print(f\"The top 3 Variance Ratios are: {explained_variance}\") "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Top 3 Eigenvectors are: [[ 1.65926472e-03 -6.81015556e-04  1.94905742e-04 -4.67130058e-03\n",
      "   1.78680075e-02  9.89829680e-04  1.56728830e-03 -1.23086662e-04\n",
      "   6.00607792e-04  2.32714319e-03  1.71380037e-04  7.04931645e-04\n",
      "   9.99822937e-01]\n",
      " [ 1.20340617e-03  2.15498184e-03  4.59369254e-03  2.64503930e-02\n",
      "   9.99344186e-01  8.77962152e-04 -5.18507284e-05 -1.35447892e-03\n",
      "   5.00440040e-03  1.51003530e-02 -7.62673115e-04 -3.49536431e-03\n",
      "  -1.77738095e-02]\n",
      " [-1.68738094e-02 -1.22003373e-01 -5.19874304e-02 -9.38593003e-01\n",
      "   2.97802484e-02  4.04846439e-02  8.54433387e-02 -1.35107797e-02\n",
      "   2.46593824e-02 -2.91398464e-01  2.59776624e-02  7.03239693e-02\n",
      "  -4.52868159e-03]]\n"
     ]
    }
   ],
   "source": [
    "print(f\"The Top 3 Eigenvectors are: {pca.components_}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "##### **Dimensionality Reduction using Linear Discriminant Analysis (LDA)**\n",
    "\n",
    "*LDA works by projecting the data onto a lower-dimensional space that maximizes the separation between the classes. It does this by finding a set of linear discriminants that maximize the ratio of between-class variance to within-class variance. In other words, it finds the directions in the feature space that best separate the different classes of data.*\n",
    "\n",
    "##### **Equations of LDA:**\n",
    "\n",
    "**1. LDA Equation for 2-Class Dataset**\n",
    "\n",
    "$$\n",
    "S_W^{-1}S_BU = \\lambda U$$\n",
    "where $\\lambda$ is the eigenvalue of the the matrix $S_W^{-1}S_B$:\n",
    "$$\n",
    "\\lambda = \\frac{{U^T}_{1\\times d}{S_B}_{d\\times d}U_{d\\times1}}{{U^T}_{1\\times d}{S_W}_{d\\times d}U_{d\\times1}}\n",
    "$$\n",
    "\n",
    "**2. LDA Equation for K-Class Dataset**\n",
    "\n",
    "$$\n",
    "S_W^{-1}S_BU = \\lambda U$$\n",
    "where $\\lambda$ is the eigenvalue of the the matrix $S_W^{-1}S_B$:\n",
    "$$\n",
    "\\lambda = \\begin{pmatrix}\n",
    "\\lambda_1 \\\\\n",
    " & \\lambda_2 &  \\\\\n",
    " & &  \\ddots & \\\\\n",
    "&&& \\lambda_n \\\\\n",
    "\\end{pmatrix}\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "\n",
    "lda_data = load_iris(as_frame=True)\n",
    "lda_features = lda_data.data.values\n",
    "lda_labels = lda_data.target.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
       "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
       "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2])"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lda_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler, LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "sc = StandardScaler()\n",
    "lda_features = sc.fit_transform(lda_features)\n",
    "le = LabelEncoder()\n",
    "lda_labels = le.fit_transform(lda_labels)\n",
    "lda_X_train, lda_X_test, lda_y_train, lda_y_test = train_test_split(lda_features, lda_labels, test_size=0.2)\n",
    " \n",
    "# apply Linear Discriminant Analysis\n",
    "lda = LinearDiscriminantAnalysis(n_components=2)\n",
    "lda_X_train_final = lda.fit_transform(lda_X_train, lda_y_train)\n",
    "lda_X_test_final = lda.transform(lda_X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Let us check the scatter plot of the first 2 components and their respective classes before Linear Discriminant Analysis (LDA)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7430f6a6fad0>"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(\n",
    "    lda_X_train[:,0],lda_X_train[:,1],c=lda_y_train,cmap='rainbow',\n",
    "  alpha=0.7,edgecolors='b'\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Now Lets see the scatterplot after using LDA to reduce the number of features to 2**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7430f6a6fc10>"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(\n",
    "    lda_X_train_final[:,0],lda_X_train_final[:,1],c=lda_y_train,cmap='rainbow',\n",
    "  alpha=0.7,edgecolors='b'\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also refer:\n",
    "\n",
    "[Dimensionality Reduction and Methods]('https://www.geeksforgeeks.org/dimensionality-reduction/')\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 3.4 Data Preprocessing - Feature Subset Selection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*What is Feature Selection? Feature Selection is the method of reducing the input variable to your model by using only relevant data and getting rid of noise in data. It is the process of automatically choosing relevant features for your machine learning model based on the type of problem you are trying to solve.*\n",
    "\n",
    "**Methods of Feature Subset Selection**:\n",
    "\n",
    "* Embedded Approaches \n",
    "    * Feature selection is done while training the model\n",
    "* Filter Approaches - Unsupervised\n",
    "    * Use only features/predictor variables\n",
    "    * Select the features that have the most information\n",
    "* Wrapper Approaches - Supervised\n",
    "    * Train using the selected subset\n",
    "    * Estimate error on the validation set \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CRIM</th>\n",
       "      <th>ZN</th>\n",
       "      <th>INDUS</th>\n",
       "      <th>CHAS</th>\n",
       "      <th>NOX</th>\n",
       "      <th>RM</th>\n",
       "      <th>AGE</th>\n",
       "      <th>DIS</th>\n",
       "      <th>TAX</th>\n",
       "      <th>PTRATIO</th>\n",
       "      <th>B</th>\n",
       "      <th>LSTAT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.00632</td>\n",
       "      <td>18.0</td>\n",
       "      <td>2.31</td>\n",
       "      <td>0</td>\n",
       "      <td>0.538</td>\n",
       "      <td>6.575</td>\n",
       "      <td>65.2</td>\n",
       "      <td>4.0900</td>\n",
       "      <td>296.0</td>\n",
       "      <td>15.3</td>\n",
       "      <td>396.90</td>\n",
       "      <td>4.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.02731</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.07</td>\n",
       "      <td>0</td>\n",
       "      <td>0.469</td>\n",
       "      <td>6.421</td>\n",
       "      <td>78.9</td>\n",
       "      <td>4.9671</td>\n",
       "      <td>242.0</td>\n",
       "      <td>17.8</td>\n",
       "      <td>396.90</td>\n",
       "      <td>9.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.02729</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.07</td>\n",
       "      <td>0</td>\n",
       "      <td>0.469</td>\n",
       "      <td>7.185</td>\n",
       "      <td>61.1</td>\n",
       "      <td>4.9671</td>\n",
       "      <td>242.0</td>\n",
       "      <td>17.8</td>\n",
       "      <td>392.83</td>\n",
       "      <td>4.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.03237</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>0</td>\n",
       "      <td>0.458</td>\n",
       "      <td>6.998</td>\n",
       "      <td>45.8</td>\n",
       "      <td>6.0622</td>\n",
       "      <td>222.0</td>\n",
       "      <td>18.7</td>\n",
       "      <td>394.63</td>\n",
       "      <td>2.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.06905</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>0</td>\n",
       "      <td>0.458</td>\n",
       "      <td>7.147</td>\n",
       "      <td>54.2</td>\n",
       "      <td>6.0622</td>\n",
       "      <td>222.0</td>\n",
       "      <td>18.7</td>\n",
       "      <td>396.90</td>\n",
       "      <td>5.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>501</th>\n",
       "      <td>0.06263</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.593</td>\n",
       "      <td>69.1</td>\n",
       "      <td>2.4786</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>391.99</td>\n",
       "      <td>9.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>502</th>\n",
       "      <td>0.04527</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.120</td>\n",
       "      <td>76.7</td>\n",
       "      <td>2.2875</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>9.08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>503</th>\n",
       "      <td>0.06076</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.976</td>\n",
       "      <td>91.0</td>\n",
       "      <td>2.1675</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>5.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>504</th>\n",
       "      <td>0.10959</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.794</td>\n",
       "      <td>89.3</td>\n",
       "      <td>2.3889</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>393.45</td>\n",
       "      <td>6.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>505</th>\n",
       "      <td>0.04741</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0.573</td>\n",
       "      <td>6.030</td>\n",
       "      <td>80.8</td>\n",
       "      <td>2.5050</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>7.88</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>506 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        CRIM    ZN  INDUS  CHAS    NOX     RM   AGE     DIS    TAX  PTRATIO  \\\n",
       "0    0.00632  18.0   2.31     0  0.538  6.575  65.2  4.0900  296.0     15.3   \n",
       "1    0.02731   0.0   7.07     0  0.469  6.421  78.9  4.9671  242.0     17.8   \n",
       "2    0.02729   0.0   7.07     0  0.469  7.185  61.1  4.9671  242.0     17.8   \n",
       "3    0.03237   0.0   2.18     0  0.458  6.998  45.8  6.0622  222.0     18.7   \n",
       "4    0.06905   0.0   2.18     0  0.458  7.147  54.2  6.0622  222.0     18.7   \n",
       "..       ...   ...    ...   ...    ...    ...   ...     ...    ...      ...   \n",
       "501  0.06263   0.0  11.93     0  0.573  6.593  69.1  2.4786  273.0     21.0   \n",
       "502  0.04527   0.0  11.93     0  0.573  6.120  76.7  2.2875  273.0     21.0   \n",
       "503  0.06076   0.0  11.93     0  0.573  6.976  91.0  2.1675  273.0     21.0   \n",
       "504  0.10959   0.0  11.93     0  0.573  6.794  89.3  2.3889  273.0     21.0   \n",
       "505  0.04741   0.0  11.93     0  0.573  6.030  80.8  2.5050  273.0     21.0   \n",
       "\n",
       "          B  LSTAT  \n",
       "0    396.90   4.98  \n",
       "1    396.90   9.14  \n",
       "2    392.83   4.03  \n",
       "3    394.63   2.94  \n",
       "4    396.90   5.33  \n",
       "..      ...    ...  \n",
       "501  391.99   9.67  \n",
       "502  396.90   9.08  \n",
       "503  396.90   5.64  \n",
       "504  393.45   6.48  \n",
       "505  396.90   7.88  \n",
       "\n",
       "[506 rows x 12 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_names = ['CRIM', 'ZN', 'INDUS', 'CHAS', 'NOX', 'RM', 'AGE', 'DIS', 'RAD', 'TAX', 'PTRATIO', 'B', 'LSTAT', 'MEDV']\n",
    "boston = pd.read_csv(filepath_or_buffer='./Datasets/Data_Preprocessing/housing.csv', header=None, delimiter=r\"\\s+\", names=column_names)\n",
    "boston = boston.drop(columns='RAD')\n",
    "boston.head()\n",
    "X = boston.drop('MEDV',axis=1)\n",
    "y = boston['MEDV']\n",
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Dropping Columns by Using Variance (Univariate Filter)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CRIM          73.986578\n",
       "ZN           543.936814\n",
       "INDUS         47.064442\n",
       "CHAS           0.064513\n",
       "NOX            0.013428\n",
       "RM             0.493671\n",
       "AGE          792.358399\n",
       "DIS            4.434015\n",
       "TAX        28404.759488\n",
       "PTRATIO        4.686989\n",
       "B           8334.752263\n",
       "LSTAT         50.994760\n",
       "MEDV          84.586724\n",
       "dtype: float64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "boston.var()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we drop the features with the least variance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CRIM</th>\n",
       "      <th>ZN</th>\n",
       "      <th>INDUS</th>\n",
       "      <th>RM</th>\n",
       "      <th>AGE</th>\n",
       "      <th>DIS</th>\n",
       "      <th>TAX</th>\n",
       "      <th>PTRATIO</th>\n",
       "      <th>B</th>\n",
       "      <th>LSTAT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.00632</td>\n",
       "      <td>18.0</td>\n",
       "      <td>2.31</td>\n",
       "      <td>6.575</td>\n",
       "      <td>65.2</td>\n",
       "      <td>4.0900</td>\n",
       "      <td>296.0</td>\n",
       "      <td>15.3</td>\n",
       "      <td>396.90</td>\n",
       "      <td>4.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.02731</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.07</td>\n",
       "      <td>6.421</td>\n",
       "      <td>78.9</td>\n",
       "      <td>4.9671</td>\n",
       "      <td>242.0</td>\n",
       "      <td>17.8</td>\n",
       "      <td>396.90</td>\n",
       "      <td>9.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.02729</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.07</td>\n",
       "      <td>7.185</td>\n",
       "      <td>61.1</td>\n",
       "      <td>4.9671</td>\n",
       "      <td>242.0</td>\n",
       "      <td>17.8</td>\n",
       "      <td>392.83</td>\n",
       "      <td>4.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.03237</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>6.998</td>\n",
       "      <td>45.8</td>\n",
       "      <td>6.0622</td>\n",
       "      <td>222.0</td>\n",
       "      <td>18.7</td>\n",
       "      <td>394.63</td>\n",
       "      <td>2.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.06905</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>7.147</td>\n",
       "      <td>54.2</td>\n",
       "      <td>6.0622</td>\n",
       "      <td>222.0</td>\n",
       "      <td>18.7</td>\n",
       "      <td>396.90</td>\n",
       "      <td>5.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>501</th>\n",
       "      <td>0.06263</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>6.593</td>\n",
       "      <td>69.1</td>\n",
       "      <td>2.4786</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>391.99</td>\n",
       "      <td>9.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>502</th>\n",
       "      <td>0.04527</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>6.120</td>\n",
       "      <td>76.7</td>\n",
       "      <td>2.2875</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>9.08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>503</th>\n",
       "      <td>0.06076</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>6.976</td>\n",
       "      <td>91.0</td>\n",
       "      <td>2.1675</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>5.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>504</th>\n",
       "      <td>0.10959</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>6.794</td>\n",
       "      <td>89.3</td>\n",
       "      <td>2.3889</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>393.45</td>\n",
       "      <td>6.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>505</th>\n",
       "      <td>0.04741</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.93</td>\n",
       "      <td>6.030</td>\n",
       "      <td>80.8</td>\n",
       "      <td>2.5050</td>\n",
       "      <td>273.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>396.90</td>\n",
       "      <td>7.88</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>506 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        CRIM    ZN  INDUS     RM   AGE     DIS    TAX  PTRATIO       B  LSTAT\n",
       "0    0.00632  18.0   2.31  6.575  65.2  4.0900  296.0     15.3  396.90   4.98\n",
       "1    0.02731   0.0   7.07  6.421  78.9  4.9671  242.0     17.8  396.90   9.14\n",
       "2    0.02729   0.0   7.07  7.185  61.1  4.9671  242.0     17.8  392.83   4.03\n",
       "3    0.03237   0.0   2.18  6.998  45.8  6.0622  222.0     18.7  394.63   2.94\n",
       "4    0.06905   0.0   2.18  7.147  54.2  6.0622  222.0     18.7  396.90   5.33\n",
       "..       ...   ...    ...    ...   ...     ...    ...      ...     ...    ...\n",
       "501  0.06263   0.0  11.93  6.593  69.1  2.4786  273.0     21.0  391.99   9.67\n",
       "502  0.04527   0.0  11.93  6.120  76.7  2.2875  273.0     21.0  396.90   9.08\n",
       "503  0.06076   0.0  11.93  6.976  91.0  2.1675  273.0     21.0  396.90   5.64\n",
       "504  0.10959   0.0  11.93  6.794  89.3  2.3889  273.0     21.0  393.45   6.48\n",
       "505  0.04741   0.0  11.93  6.030  80.8  2.5050  273.0     21.0  396.90   7.88\n",
       "\n",
       "[506 rows x 10 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = X.drop(['NOX','CHAS'],axis=1)\n",
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Dropping features based on Correlation (Univariate Filters)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig_dims = (12, 8)\n",
    "fig, ax = plt.subplots(figsize=fig_dims)\n",
    "sns.heatmap(boston.corr(), ax=ax)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CRIM       0.388305\n",
       "ZN         0.360445\n",
       "INDUS      0.483725\n",
       "CHAS       0.175260\n",
       "NOX        0.427321\n",
       "RM         0.695360\n",
       "AGE        0.376955\n",
       "DIS        0.249929\n",
       "TAX        0.468536\n",
       "PTRATIO    0.507787\n",
       "B          0.333461\n",
       "LSTAT      0.737663\n",
       "MEDV       1.000000\n",
       "Name: MEDV, dtype: float64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "abs(boston.corr()[\"MEDV\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The list of columns that are below the threshold of 0.3 are: ['CHAS', 'DIS']\n"
     ]
    }
   ],
   "source": [
    "print(f\"The list of columns that are below the threshold of 0.3 are: {abs(boston.corr()['MEDV'][abs(boston.corr()['MEDV'])<0.3]).index.to_list()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Chi-Squared Test - Filter Feature Selection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Chi-Squared Statistic ($X^2$) is given by:\n",
    "$$ X^2 = \\Sigma\\frac{(O-E)^2}{E} $$\n",
    "\n",
    "where $O$ is the contingency table and $E$ is:\n",
    "$$ E = \\frac{\\Sigma_{row}*\\Sigma_{col}}{N} $$\n",
    "\n",
    "**If  $X^2$ >= $z_\\alpha$, then reject $H_0$, else accept $H_0$. Similarly, if $X^2$ >= $z_\\alpha$, then accept $H_A$, else reject $H_A$.**  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "tips=sns.load_dataset('tips')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Null Hypothesis($H_0$): Smoking Status is Independent of Gender**\n",
    "\n",
    "**Alternate Hypothesis($H_A$): Smoking Status is Dependent on Gender**\n",
    "\n",
    "**$\\alpha = 0.5$**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[60, 97],\n",
       "       [33, 54]])"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "contingency_table = pd.crosstab(tips['sex'],tips['smoker'])\n",
    "contingency_table.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import chi2_contingency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[59.84016393, 97.15983607],\n",
       "       [33.15983607, 53.84016393]])"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "_,_,_, expected_freq = chi2_contingency(contingency_table)\n",
    "expected_freq"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Degrees of Freedom: 1\n"
     ]
    }
   ],
   "source": [
    "no_of_rows=len(contingency_table.iloc[0:2,0])\n",
    "no_of_columns=len(contingency_table.iloc[0,0:2])\n",
    "ddof=(no_of_rows-1)*(no_of_columns-1)\n",
    "print(\"Degrees of Freedom:\",ddof)\n",
    "alpha = 0.05"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Chi-Squared Statistic is: 0.001934818536627623\n",
      "The Critical Value is: 3.841458820694124\n",
      "The p-value of the test is: 0.964915107315732\n"
     ]
    }
   ],
   "source": [
    "from scipy.stats import chi2\n",
    "chi_square=sum([(o-e)**2./e for o,e in zip(contingency_table.values,expected_freq)])\n",
    "chi_square_statistic=chi_square[0]+chi_square[1]\n",
    "     \n",
    "\n",
    "print(\"The Chi-Squared Statistic is:\",chi_square_statistic)\n",
    "critical_value=chi2.ppf(q=1-alpha,df=ddof)\n",
    "print('The Critical Value is:',critical_value)\n",
    "p_value=1-chi2.cdf(x=chi_square_statistic,df=ddof)\n",
    "print(f\"The p-value of the test is: {p_value}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since Chi Square Statistic > Critical Value:\n",
    " \n",
    "**Reject $H_0$, There is a dependence between the two variables**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Using Wrapper Selection (Sequential Feature Selection) for a Logistic Regression Model\n",
    "\n",
    "*Sequential feature selection (SFS) is a greedy algorithm that iteratively adds or removes features from a dataset in order to improve the performance of a predictive model. SFS can be either forward selection or backward selection. Forward selection is a step-wise feature selection method that starts with an empty feature set and iteratively adds one feature at a time based on certain evaluation criteria.Backward selection is the opposite of forward-select where you start with a full feature set and keep eliminating features.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.feature_selection import SequentialFeatureSelector"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "iris = load_iris(as_frame=True)\n",
    "X = iris.data\n",
    "y = iris.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The selected features are: ['petal length (cm)', 'petal width (cm)']\n"
     ]
    }
   ],
   "source": [
    "logreg = LogisticRegression()\n",
    " \n",
    "# Create a sequential feature selector\n",
    "selector = SequentialFeatureSelector(\n",
    "    logreg, n_features_to_select=2, scoring='accuracy')\n",
    " \n",
    "# Fit the selector to the data\n",
    "selector.fit(X, y)\n",
    " \n",
    "# Get the selected features\n",
    "selected_features = selector.get_support()\n",
    " \n",
    "print('The selected features are:', list(X.columns[selected_features]))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also refer:\n",
    "\n",
    "[Wrapper-Based Feature Selection]('https://www.analyticsvidhya.com/blog/2020/10/a-comprehensive-guide-to-feature-selection-using-wrapper-methods-in-python/')\n",
    "\n",
    "[Filter-Based Feature Selection]('https://medium.com/analytics-vidhya/feature-selection-73bc12a9b39e')\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 3.5 Data Preprocessing - Feature Creating Discretization and Binarization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Dealing with Continuous Data\n",
    "\n",
    "##### **Discretization** \n",
    "*Data discretization refers to a method of converting a huge number of data values into smaller ones so that the evaluation and management of data become easy. In other words, data discretization is a method of converting attributes values of continuous data into a finite set of intervals with minimum data loss.*\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ID</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Age\n",
       "ID     \n",
       "1    19\n",
       "2    70\n",
       "3    36\n",
       "4    13\n",
       "5    42"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "discretization_df = pd.read_csv('./Datasets/Data_Preprocessing/discretization.csv',index_col='ID')\n",
    "discretization_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The number of unique age values are: 21\n"
     ]
    }
   ],
   "source": [
    "print(f\"The number of unique age values are: {len(discretization_df['Age'].unique())}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the above dataset, we can discretize the age into fewer values by adding some thresholds:\n",
    "* If $age>=0$ and $age<13$, then the data can be categorized as **Child**\n",
    "* If $age>=13$ and $age<20$, then the data can be categorized as **Teenager**\n",
    "* If $age>=20$ and $age<60$, then the data can be categorized as **Adult**\n",
    "* If $age>=60$ and $age<110$, then the data can be categorized as **Old**\n",
    "* If $age>=110$ or $age<0$, the the data can be categorized as **Invalid**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Categorical</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ID</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Teenager</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Old</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Adult</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Teenager</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Adult</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Categorical\n",
       "ID            \n",
       "1     Teenager\n",
       "2          Old\n",
       "3        Adult\n",
       "4     Teenager\n",
       "5        Adult"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "col_values = ['Child','Teenager','Adult','Old','Invalid']\n",
    "def get_category(age):\n",
    "    if age>= 0 and age<13:\n",
    "        return col_values[0]\n",
    "    elif age>=13 and age<20:\n",
    "        return col_values[1]\n",
    "    elif age>=20 and age<60:\n",
    "        return col_values[2]\n",
    "    elif age>=60 and age<110:\n",
    "        return col_values[3]\n",
    "    elif age>=110 or age<0:\n",
    "        return col_values[4]\n",
    "discretization_df['Categorical'] = discretization_df['Age'].apply(get_category)\n",
    "discretization_df = discretization_df.drop('Age',axis=1)\n",
    "discretization_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The number of unique age values are: ['Teenager' 'Old' 'Adult' 'Child']\n"
     ]
    }
   ],
   "source": [
    "print(f\"The number of unique age values are: {discretization_df['Categorical'].unique()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Binarization\n",
    "\n",
    "##### **Approach 1: Representing category in binary format**\n",
    "The process of binarizing involves converting this newly formed categorical data into numbers that can be fed into statistical models. If the categorical column contains $m$ categories then we add $\\log_2{m}$ columns to the data and represent the categories in binary number representation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Categorical</th>\n",
       "      <th>x1</th>\n",
       "      <th>x2</th>\n",
       "      <th>x3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Teenager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Old</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Adult</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Teenager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Adult</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Teenager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Child</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Adult</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Old</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Teenager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Adult</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Old</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Old</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Teenager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Child</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Child</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Adult</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Adult</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Child</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Child</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Child</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Categorical  x1  x2  x3\n",
       "ID                        \n",
       "1     Teenager   0   0   0\n",
       "2          Old   0   0   0\n",
       "3        Adult   0   0   0\n",
       "4     Teenager   0   0   0\n",
       "5        Adult   0   0   0\n",
       "6     Teenager   0   0   0\n",
       "7        Child   0   0   0\n",
       "8        Adult   0   0   0\n",
       "9          Old   0   0   0\n",
       "10    Teenager   0   0   0\n",
       "11       Adult   0   0   0\n",
       "12         Old   0   0   0\n",
       "13         Old   0   0   0\n",
       "14    Teenager   0   0   0\n",
       "15       Child   0   0   0\n",
       "16       Child   0   0   0\n",
       "17       Adult   0   0   0\n",
       "18       Adult   0   0   0\n",
       "19       Child   0   0   0\n",
       "20       Child   0   0   0\n",
       "21       Child   0   0   0"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for i in range(math.floor(math.log2(len(discretization_df['Categorical'].unique())))+1):\n",
    "    \n",
    "    discretization_df[f\"x{i+1}\"] = 0\n",
    "\n",
    "discretization_df\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "discretization_df['Categorical'][1] =='Teenager'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Categorical</th>\n",
       "      <th>x1</th>\n",
       "      <th>x2</th>\n",
       "      <th>x3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Teenager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Old</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Adult</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Teenager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Adult</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Categorical  x1  x2  x3\n",
       "ID                        \n",
       "1     Teenager   0   0   1\n",
       "2          Old   0   1   1\n",
       "3        Adult   0   1   0\n",
       "4     Teenager   0   0   1\n",
       "5        Adult   0   1   0"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%capture --no-display\n",
    "for i in range(1,discretization_df['Categorical'].count()+1):\n",
    "    if discretization_df['Categorical'][i] == 'Teenager': \n",
    "        discretization_df['x1'][i] = 0\n",
    "        discretization_df['x2'][i] = 0\n",
    "        discretization_df['x3'][i] = 1\n",
    "    elif discretization_df['Categorical'][i] == 'Adult': \n",
    "        discretization_df['x1'][i] = 0\n",
    "        discretization_df['x2'][i] = 1\n",
    "        discretization_df['x3'][i] = 0\n",
    "    elif discretization_df['Categorical'][i] == 'Child': \n",
    "        discretization_df['x1'][i] = 0\n",
    "        discretization_df['x2'][i] = 0\n",
    "        discretization_df['x3'][i] = 0\n",
    "    elif discretization_df['Categorical'][i] == 'Old': \n",
    "        discretization_df['x1'][i] = 0\n",
    "        discretization_df['x2'][i] = 1\n",
    "        discretization_df['x3'][i] = 1\n",
    "\n",
    "discretization_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Approach 2: Representing categories as one-hot encoded values**\n",
    "\n",
    "In this approach if there are $m$ categories, we add $m$ columns to the dataset and one-hot encode the categories. For Association related problems, this approach is required because association looks only for the precense of the attribute ($value=1$) in the table "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelBinarizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "lb = LabelBinarizer().fit(discretization_df['Categorical'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Categorical</th>\n",
       "      <th>x1_x</th>\n",
       "      <th>x2_x</th>\n",
       "      <th>x3_x</th>\n",
       "      <th>x1_y</th>\n",
       "      <th>x2_y</th>\n",
       "      <th>x3_y</th>\n",
       "      <th>x4</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Teenager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Old</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Adult</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Teenager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Adult</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Categorical  x1_x  x2_x  x3_x  x1_y  x2_y  x3_y  x4\n",
       "ID                                                    \n",
       "1     Teenager     0     0     1     0     0     0   1\n",
       "2          Old     0     1     1     0     0     1   0\n",
       "3        Adult     0     1     0     1     0     0   0\n",
       "4     Teenager     0     0     1     0     0     0   1\n",
       "5        Adult     0     1     0     1     0     0   0"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "binarized_df = pd.DataFrame(np.hstack([discretization_df.index.to_numpy().reshape(-1,1),lb.transform(discretization_df['Categorical'])]),columns=['ID','x1','x2','x3','x4'])\n",
    "binarized_df = binarized_df.set_index('ID')\n",
    "binarized_df = discretization_df.merge(binarized_df,'inner',left_index=True,right_index=True)\n",
    "binarized_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also refer:\n",
    "\n",
    "[Discretization in Data Mining]('https://www.javatpoint.com/discretization-in-data-mining')\n",
    "\n",
    "[Binarization and Implementation in Python]('https://www.tutorialspoint.com/how-to-binarize-the-data-using-python-scikit-learn')\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 3.6 Data Preprocessing - Attribute Transformation\n",
    "\n",
    "*Attribute transformation alters the data by replacing a selected attribute by one or more new attributes, functionally dependent on the original one, to facilitate further analysis*\n",
    "\n",
    "\n",
    "\n",
    "    Note: In Data Mining, the terms Standardization and Normalization are used interchangeably. This is in contrast to ML Jargon where Standardization and Normalization are two different approaches."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Methods of Attribute Transformation\n",
    "\n",
    "**Approach 1: Standardization or Normalization**\n",
    "\n",
    "In this approach we scale the data using the formula:\n",
    "$$ X_{new} = \\frac{X-\\bar{X}}{\\sigma} $$\n",
    "\n",
    "This approach is preferred when the features are of different scales"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>20.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.023809</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.045339</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-1.196154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-0.816165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-0.256183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.543793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.883751</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Age\n",
       "count  20.000000\n",
       "mean    0.023809\n",
       "std     1.045339\n",
       "min    -1.196154\n",
       "25%    -0.816165\n",
       "50%    -0.256183\n",
       "75%     0.543793\n",
       "max     1.883751"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaling_df = pd.read_csv('./Datasets/Data_Preprocessing/discretization.csv',index_col='ID')\n",
    "standard_scaler = StandardScaler()\n",
    "standard_scaler.fit(scaling_df)\n",
    "scaled_data = standard_scaler.transform(scaling_df)\n",
    "scaling_df['Age'] = pd.DataFrame(scaled_data,columns = ['Scaled'])['Scaled']\n",
    "scaling_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can observe, the data above has been scaled to a normal distribution which is mean centered and has a standard distribution of 1."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Approach 2: Min Max Normalization** \n",
    "\n",
    "In this approach, we transform the data using the formula:\n",
    "$$ X_{new} = \\frac{X-X_{min}}{X_{max}-X_{min}} $$\n",
    "\n",
    "This approach can scale down any data into the ranges [-1,1] which makes it very useful for such cases"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>20.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.396104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.339406</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.123377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.305195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.564935</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Age\n",
       "count  20.000000\n",
       "mean    0.396104\n",
       "std     0.339406\n",
       "min     0.000000\n",
       "25%     0.123377\n",
       "50%     0.305195\n",
       "75%     0.564935\n",
       "max     1.000000"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaling_df = pd.read_csv('./Datasets/Data_Preprocessing/discretization.csv',index_col='ID')\n",
    "minmax_scaler = MinMaxScaler()\n",
    "scaled_data = minmax_scaler.fit_transform(scaling_df)\n",
    "scaling_df['Age'] = pd.DataFrame(scaled_data,columns = ['Scaled'])['Scaled']\n",
    "scaling_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also refer:\n",
    "\n",
    "[Data Transformation in Data Mining]('https://www.geeksforgeeks.org/data-transformation-in-data-mining/')\n",
    "\n",
    "---"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}